v5.27 (MB) live on main.

Message boards : Number crunching : v5.27 (MB) live on main.
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · 6 . . . 9 · Next

AuthorMessage
Profile Gavin Shaw
Avatar

Send message
Joined: 8 Aug 00
Posts: 1116
Credit: 1,304,337
RAC: 0
Australia
Message 617408 - Posted: 10 Aug 2007, 5:13:31 UTC - in response to Message 617394.  

Sorry to burst your bubble but did you check the date on that wu? It was not a mb wu. You did get the mb application, but your work unit is from 2000, definitely not a multibeam!

Is there a quick and easy way to tell?

I've got a fresh one that appears to be data logged on March 3, 2007, and which has the impressively far in the future return deadline of October 1 (over 50 days from date of issue). Is this perhaps a multibeam? If not, how to tell?
perhaps a multibeam WU?


I'm not sure without doing some research the exact date that the multibeam recorder recorded it's first work, so I can't say an exact date to look for but yours and Byron's both are one of the multibeam. You can tell by the date as an easy way without reading into header files and such. The date is the first three groups of two characters in the filename in the format ddmmyy. In this case you both have one that was recorded on the third of March, 2007. Most of the linefeed work we are running now has dates in the 1999 to 2001 range.


To tell if it is MB, just look at the graphics (if you have the 5.27 stock app). Those that are MB say that. Those that are not, just say 'Enhanced'.

Never surrender and never give up. In the darkest hour there is always hope.

ID: 617408 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 617409 - Posted: 10 Aug 2007, 5:17:00 UTC - in response to Message 617407.  


... If properly done, a shutdown followed by a reset should cause the work units to be resent as "lost work units" and the new application downloaded to crunch them.


Thanks Jim, quite a while before I get to that wu,[ perhaps a full day] will give me time to think about doing just that, or maybe there'll be a chicken update first, fingers crossed.



"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 617409 · Report as offensive
Profile Jim-R.
Volunteer tester
Avatar

Send message
Joined: 7 Feb 06
Posts: 1494
Credit: 194,148
RAC: 0
United States
Message 617410 - Posted: 10 Aug 2007, 5:17:43 UTC - in response to Message 617408.  



To tell if it is MB, just look at the graphics (if you have the 5.27 stock app). Those that are MB say that. Those that are not, just say 'Enhanced'.

Yes, but this only works for a wu that is already started crunching. You can't look at any wu in the cache with this technique. Checking the date on the wu's filename will tell you for any wu in the cache.
Jim

Some people plan their life out and look back at the wealth they've had.
Others live life day by day and look back at the wealth of experiences and enjoyment they've had.
ID: 617410 · Report as offensive
Profile Jim-R.
Volunteer tester
Avatar

Send message
Joined: 7 Feb 06
Posts: 1494
Credit: 194,148
RAC: 0
United States
Message 617411 - Posted: 10 Aug 2007, 5:21:28 UTC - in response to Message 617409.  
Last modified: 10 Aug 2007, 5:27:13 UTC


... If properly done, a shutdown followed by a reset should cause the work units to be resent as "lost work units" and the new application downloaded to crunch them.


Thanks Jim, quite a while before I get to that wu,[ perhaps a full day] will give me time to think about doing just that, or maybe there'll be a chicken update first, fingers crossed.



If it's going to be that long before you start it, just set "no new work" now and let it crunch the task already running. That way the only one you will have the chance to loose is the one wu. Of course this is assuming that the wu in question was the last downloaded, or you only have the two wu's. Matter of fact, if you do start crunching on it before the rest of them are crunched just suspend it and let it finish the others, then make the switch. The only time you will loose is the amount of crunching it does on it before you suspend it (assuming it doesn't error out immediately that is!). Good luck.
Jim

Some people plan their life out and look back at the wealth they've had.
Others live life day by day and look back at the wealth of experiences and enjoyment they've had.
ID: 617411 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 617412 - Posted: 10 Aug 2007, 5:33:17 UTC - in response to Message 617411.  


If it's going to be that long before you start it, just set "no new work" now and let it crunch the task already running. That way the only one you will have the chance to loose is the one wu. Of course this is assuming that the wu in question was the last downloaded, or you only have the two wu's. Matter of fact, if you do start crunching on it before the rest of them are crunched just suspend it and let it finish the others, then make the switch. The only time you will loose is the amount of crunching it does on it before you suspend it (assuming it doesn't error out immediately that is!). Good luck.


Well it's about fourth from last of about 19 tasks, so I have set No New Tasks, and suspended that particular multibeam one, so I guess it should sit there until last, then I'll try the app_info.xml delete and reset.
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 617412 · Report as offensive
PEWD

Send message
Joined: 13 Jun 07
Posts: 7
Credit: 136,121
RAC: 0
United States
Message 617415 - Posted: 10 Aug 2007, 5:36:33 UTC

hi - this is my first post, i'm not sure i have it in the correct place.

I'm seeing this on several recent work units, anyone know what this error msg means?

tia.

<core_client_version>5.10.7</core_client_version>
<![CDATA[
<stderr_txt>
setiathome_enhanced 5.27 DevC++/MinGW

Work Unit Info:
...............
WU true angle range is : 11.401485
Optimal function choices:
-----------------------------------------------------
name
-----------------------------------------------------
v_BaseLineSmooth (no other)
v_vGetPowerSpectrumUnrolled 0.00073 0.00000
sse1_ChirpData_ak 0.01883 0.00000
v_vTranspose4x16ntw 0.00622 0.00000
BH SSE folding 0.00158 0.00000
SETI@Home Informational message -9 result_overflow
NOTE: The number of results detected exceeds the storage space allocated.

Flopcounter: 2620179655.273146
ID: 617415 · Report as offensive
Alinator
Volunteer tester

Send message
Joined: 19 Apr 05
Posts: 4178
Credit: 4,647,982
RAC: 0
United States
Message 617416 - Posted: 10 Aug 2007, 5:43:14 UTC
Last modified: 10 Aug 2007, 5:43:45 UTC

It's not an error per se. It's what is commonly referred to as a 'noisey' WU, and just means more signals are found than is allowed to be stored in the output data file.

HTH,

Alinator
ID: 617416 · Report as offensive
W-K 666 Project Donor
Volunteer tester

Send message
Joined: 18 May 99
Posts: 19359
Credit: 40,757,560
RAC: 67
United Kingdom
Message 617417 - Posted: 10 Aug 2007, 5:47:41 UTC - in response to Message 617415.  

hi - this is my first post, i'm not sure i have it in the correct place.

I'm seeing this on several recent work units, anyone know what this error msg means?

tia.

<core_client_version>5.10.7</core_client_version>
<![CDATA[
<stderr_txt>
setiathome_enhanced 5.27 DevC++/MinGW

Work Unit Info:
...............
WU true angle range is : 11.401485
Optimal function choices:
-----------------------------------------------------
name
-----------------------------------------------------
v_BaseLineSmooth (no other)
v_vGetPowerSpectrumUnrolled 0.00073 0.00000
sse1_ChirpData_ak 0.01883 0.00000
v_vTranspose4x16ntw 0.00622 0.00000
BH SSE folding 0.00158 0.00000
SETI@Home Informational message -9 result_overflow
NOTE: The number of results detected exceeds the storage space allocated.

Flopcounter: 2620179655.273146

This error msg is quite normal, it is produced when too many signals are found, there is only so much room on the database. Usually when a high number of signals are detected in a unit it is usually from man made noise, radio interference.
You should find that your partner for that unit gets the same -9 error.
If your partner doesn't and you get this situation on a regular basis then it can also be a sign that your computer has become unstable, usually overheating. But this is fairly rare.
ID: 617417 · Report as offensive
Alinator
Volunteer tester

Send message
Joined: 19 Apr 05
Posts: 4178
Credit: 4,647,982
RAC: 0
United States
Message 617419 - Posted: 10 Aug 2007, 5:51:30 UTC - in response to Message 617407.  
Last modified: 10 Aug 2007, 5:52:59 UTC

Okay, So I have one of these '07 units sitting in the cache of one of my machines. My questions are:
- I am running chicken 2.2B, will it run/validate correctly?
- if it runs, will it run longer or faster than stock 5.27?
- will it give me more or less credit/time than stock 5.27?
Jason

MB wu http://setiathome.berkeley.edu/workunit.php?wuid=146173344

As posted by Papa, there have been numerous problems running the mb units with Chicken's app. Many have run and crunched fine, but many more have not. So it may not validate, or may give other errors. However at this point unless you are willing to do a project reset and take the chance of losing them both you will just have to take your chances with it. I would set "no new work" until you can crunch your cache down and remove your app_info.xml file and let the system download the new app.
If properly done, a shutdown followed by a reset should cause the work units to be resent as "lost work units" and the new application downloaded to crunch them.


Wait a second here. You guys got some links to hosts producing a lot of validate failures of the Coop app on MB data over on Beta? It's not that I don't believe you, it's just that I'm from Missouri. ;-)

I went back and looked over CElliot's host #5903 (the only one I remember offhand), and didn't see any signs of it having any trouble running 2.0 and 2.2B on the MB data while he was running it that way. We're talking around a couple thousand results run this way.

Alinator
ID: 617419 · Report as offensive
P . P . L .
Volunteer tester

Send message
Joined: 7 Jun 03
Posts: 86
Credit: 161,216
RAC: 0
Australia
Message 617428 - Posted: 10 Aug 2007, 6:17:54 UTC

G'day all.

Joe or someone else in the know, can i use the 5.27app from Beta or has the

credit claimed changed between here and there again.

Thanks.

ID: 617428 · Report as offensive
Profile Jim-R.
Volunteer tester
Avatar

Send message
Joined: 7 Feb 06
Posts: 1494
Credit: 194,148
RAC: 0
United States
Message 617437 - Posted: 10 Aug 2007, 7:09:18 UTC - in response to Message 617419.  


Wait a second here. You guys got some links to hosts producing a lot of validate failures of the Coop app on MB data over on Beta? It's not that I don't believe you, it's just that I'm from Missouri. ;-)

I went back and looked over CElliot's host #5903 (the only one I remember offhand), and didn't see any signs of it having any trouble running 2.0 and 2.2B on the MB data while he was running it that way. We're talking around a couple thousand results run this way.

Alinator

You're so right, the problems I was refering to was for the win95/98 support.

Jason..

I apologize for the mistake. You can crunch right along. The reference to Papa's message refering to the optimized app led me to think about a Windows problem, however the Windows problem is with 95/98 etc. (Half asleep when I read that and been reading too many forum posts for one night! Hehe)
Jim

Some people plan their life out and look back at the wealth they've had.
Others live life day by day and look back at the wealth of experiences and enjoyment they've had.
ID: 617437 · Report as offensive
Tin Man

Send message
Joined: 18 Sep 06
Posts: 15
Credit: 1,623,138
RAC: 0
United Kingdom
Message 617440 - Posted: 10 Aug 2007, 7:38:34 UTC
Last modified: 10 Aug 2007, 7:55:39 UTC

Hi all
I think i've just crunched my first mb wu. however i'm i little confused. It was crunched using seti 5.17 although according to my seti folder i've already got v 5.27 downloaded. it also used simons chicken app and showed up as a seti enhanced application whist crunching i've deleted the info.app xml several hours ago Is all ok or do i have to do something else to get 5.27 to run
ID: 617440 · Report as offensive
W-K 666 Project Donor
Volunteer tester

Send message
Joined: 18 May 99
Posts: 19359
Credit: 40,757,560
RAC: 67
United Kingdom
Message 617464 - Posted: 10 Aug 2007, 9:35:33 UTC - in response to Message 617440.  

Hi all
I think i've just crunched my first mb wu. however i'm i little confused. It was crunched using seti 5.17 although according to my seti folder i've already got v 5.27 downloaded. it also used simons chicken app and showed up as a seti enhanced application whist crunching i've deleted the info.app xml several hours ago Is all ok or do i have to do something else to get 5.27 to run

You have to clear your work queue, using no new tasks, and then stop and restart BOINC, then allow new tasks. Whilst you still have work using 5.17 on your list it will keep using it.

Andy
ID: 617464 · Report as offensive
Idefix
Volunteer tester

Send message
Joined: 7 Sep 99
Posts: 154
Credit: 482,193
RAC: 0
Germany
Message 617468 - Posted: 10 Aug 2007, 10:04:45 UTC - in response to Message 617408.  
Last modified: 10 Aug 2007, 10:05:36 UTC

Hi,
Is there a quick and easy way to tell?

To tell if it is MB, just look at the graphics (if you have the 5.27 stock app). Those that are MB say that. Those that are not, just say 'Enhanced'.

Or look at the name of the WU.

"classic" enhanced:
31mr00aa.9721.15760.1028384.3.230

multi beam enhanced:
03mr07aa.15818.6207.4.4.142

And here you can see the beam and polarisation:
03mr07aa.15818.6207.4.4.142

3 --> beam 0, pol 0
4 --> beam 0, pol 1
5 --> beam 1, pol 0
6 --> beam 1, pol 1
and so on

Regards,
Carsten
ID: 617468 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 617471 - Posted: 10 Aug 2007, 10:33:44 UTC - in response to Message 617437.  


You're so right, the problems I was refering to was for the win95/98 support.

Jason..

I apologize for the mistake. You can crunch right along. The reference to Papa's message refering to the optimized app led me to think about a Windows problem, however the Windows problem is with 95/98 etc. (Half asleep when I read that and been reading too many forum posts for one night! Hehe)


All good Jim :D, have only set new tasks so I can let it do the switch later anyway, I'll unsuspend that MB task and let it crunch dry, Cheers !
"Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions.
ID: 617471 · Report as offensive
Dominik S.

Send message
Joined: 4 Jun 03
Posts: 15
Credit: 4,346,294
RAC: 0
Poland
Message 617478 - Posted: 10 Aug 2007, 11:06:22 UTC - in response to Message 617468.  

Hi,
Is there a quick and easy way to tell?

To tell if it is MB, just look at the graphics (if you have the 5.27 stock app). Those that are MB say that. Those that are not, just say 'Enhanced'.

Or look at the name of the WU.

"classic" enhanced:
31mr00aa.9721.15760.1028384.3.230

multi beam enhanced:
03mr07aa.15818.6207.4.4.142

And here you can see the beam and polarisation:
03mr07aa.15818.6207.4.4.142

3 --> beam 0, pol 0
4 --> beam 0, pol 1
5 --> beam 1, pol 0
6 --> beam 1, pol 1
and so on

Regards,
Carsten

Or look into WU (this is text file) and search for string <receiver_cfg>
after that string you can find string <name>Arecibo 1.4GHz Array, Beam 0, Pol 0</name> in MB WU or string <name>ao1420</name> in old enhanced WU

ID: 617478 · Report as offensive
Tin Man

Send message
Joined: 18 Sep 06
Posts: 15
Credit: 1,623,138
RAC: 0
United Kingdom
Message 617488 - Posted: 10 Aug 2007, 11:16:10 UTC - in response to Message 617464.  

Hi all
I think i've just crunched my first mb wu. however i'm i little confused. It was crunched using seti 5.17 although according to my seti folder i've already got v 5.27 downloaded. it also used simons chicken app and showed up as a seti enhanced application whist crunching i've deleted the info.app xml several hours ago Is all ok or do i have to do something else to get 5.27 to run

You have to clear your work queue, using no new tasks, and then stop and restart BOINC, then allow new tasks. Whilst you still have work using 5.17 on your list it will keep using it.

Andy



Thanks andy
ID: 617488 · Report as offensive
archae86

Send message
Joined: 31 Aug 99
Posts: 909
Credit: 1,582,816
RAC: 0
United States
Message 617505 - Posted: 10 Aug 2007, 12:08:09 UTC - in response to Message 617468.  
Last modified: 10 Aug 2007, 12:16:54 UTC

Is there a quick and easy way to tell?

Or look at the name of the WU.

"classic" enhanced:
31mr00aa.9721.15760.1028384.3.230

multi beam enhanced:
03mr07aa.15818.6207.4.4.142

And here you can see the beam and polarisation:
03mr07aa.15818.6207.4.4.142

3 --> beam 0, pol 0
4 --> beam 0, pol 1
5 --> beam 1, pol 0
6 --> beam 1, pol 1
and so on

Regards,
Carsten

Thanks Carsten, I like that way. Two single-digit period-delimited fields just before the last field to the right seems simple enough to look for.

I'm bemused by the October 1 reply deadline, by the way. I feared this might be a monster, but my E6600 core 2 Duo (running 3 GHz on a moderate overclock) thinks it is 70% done with only an hour and a half of CPU time invested.

It makes me wonder whether the variable deadline function has not been calibrated for Multibeam to anywhere close to the previous work.

ID: 617505 · Report as offensive
Josef W. Segur
Volunteer developer
Volunteer tester

Send message
Joined: 30 Oct 99
Posts: 4504
Credit: 1,414,761
RAC: 0
United States
Message 617539 - Posted: 10 Aug 2007, 15:20:06 UTC - in response to Message 617428.  

G'day all.

Joe or someone else in the know, can i use the 5.27app from Beta or has the

credit claimed changed between here and there again.

Thanks.

The 5.27 here is identical to the 5.27 from Beta.
                                                              Joe
ID: 617539 · Report as offensive
Josef W. Segur
Volunteer developer
Volunteer tester

Send message
Joined: 30 Oct 99
Posts: 4504
Credit: 1,414,761
RAC: 0
United States
Message 617544 - Posted: 10 Aug 2007, 15:45:35 UTC - in response to Message 617505.  

...
I'm bemused by the October 1 reply deadline, by the way. I feared this might be a monster, but my E6600 core 2 Duo (running 3 GHz on a moderate overclock) thinks it is 70% done with only an hour and a half of CPU time invested.

It makes me wonder whether the variable deadline function has not been calibrated for Multibeam to anywhere close to the previous work.

The splitter calculations on which time estimates are based were developed in 2005 during the early phase of setiathome_enhanced development. There were a few adjustments made subsequently to shift sections up or down, but nothing to change the basic shapes. Eric is aware the curves don't match reality, just hasn't had time to change them.

These 0.405 angle range WUs are estimated quite long. The curve for the 0.226 through 1.127 angle ranges where Gaussian searches are performed is much steeper than needed. Near 1.127 the estimates are far less generous.
                                                               Joe
ID: 617544 · Report as offensive
Previous · 1 · 2 · 3 · 4 · 5 · 6 . . . 9 · Next

Message boards : Number crunching : v5.27 (MB) live on main.


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.